Over 10,000 GB can be stored in this tiny pink droplet! DNA storage a possibility? UW and Microsoft partnered up to create a method of accurately storing and recovering hard drive data into DNA snippets. Their latest trial was perfect at recovering data due to their new approach to encryption and decryption. (via University of Washington)
Wetware on the way?
Microsoft Research has currently decided to change the market for archival data storage by utilizing DNA to store millions of gigabytes of data in a single gram of DNA. However, in order to achieve this feat, which we recently posted about, they have teamed up with some researchers in the University of Washington; they shared their findings in a paper presented at ACM International Conference on Architectural Support for Programming Languages and Operating Systems.
Their paper elaborates on how Microsoft Research labs have been able to successfully store and retrieve data encoded in synthetic DNA with the help of a collaboration with University of Washington researchers. So far, this team is one of only two researchers to successfully encode and retrieve data stored in DNA with a one hundred percent success rate.
So, what’s the secret? The secret seems to lie in the encoding and decryption process. The process used to create and read the DNA is fairly simple. Once they have encoded a chunk of data into letters A, C, G, and T: the nucleotides which are the building blocks for DNA. They then outsource the creation of spinets of DNA strands which utilize their encoded sequence of letters.
To retrieve the data, they must sequence the DNA strands which are all together in the same test tube (seen above as a tiny speck of pink). So, of course, the decoding process is more involved than simply finding out the sequencing of the DNA within the test tube: you have to decode it. And here is where this team up of interdisciplinary scientists from Microsoft and University of Washington got it very right!
They put the magic into how they chose to encode the data from it’s original bits of zeros and ones into nucleotides A, C, G, and T. They knew that, if they could streamline their process, they would have little to no errors later in the decryption process. Essentially, they tried to make it as streamlined and simple as possible to avoid the errors that come with complexity. But how could they know where each snippet of DNA fell in the full sequence of the data? They encoded zip codes and street address equivalents into each snippet of DNA to correctly place each sequence into the bigger sequence for accurate decryption. A pretty clever and simple solution, right?
All in all, their novel approach to encryption and decryption paid off as they were able to restore all of the data from the DNA without any errors or data loss. The whole project is impressive, but this current method can only work for storing archival data that requires no alterations and no immediate access. While this can provide a good service to companies who have large data stores of information, I wonder how practical this really is. On the one hand, one drop of DNA can store about 10,000 GBs. On the other hand, what is our obsession with storing everything?!
This can also present a sort of breach of security as companies like Facebook will have a copy of all of your photos and your profile for eternity – long after you choose to delete your profile and cancel your account? Also, with the new compactness of DNA data storage will companies choose to keep archival data forever, rather than for 5-10 years when they run out of hard disk space? Where is the line drawn, and what are the rights of customers if their archival data (which could include SSN and bank information) is stored forever by a company that they no longer choose to actively do business with?
Have a story tip? Message me at: